The paper investigates the relationship between some car styles and their images using fuzzy set theory. Fifteen car styles with very different characteristics and ten image words are selected to analyse the image sensation of each style. The person's image perceptionof a car is quantified using the membership function. A computer program is constructed to calculate and analyse the similarity relations of various styles and further to arrange the corresponding membership functions of styles to one image word by size. The results showthat a car formed with smooth curved surfaces all over the body enjoys wide appeal. The ordered table can be used to suggest what stylea designer should choose if he inputs the required image word (or words) to specify the characteristics of a car at the concept stage. In this way, not only can design decision errors be reduced, the efficiency of the decision making can also be improved. It is thus a powerful and variable method for designing a product
Abstract